Article: Identifying diagnosis and mortality of COVID-19 by learning a sequence-to-sequence ARIMA-based model Journal: International Journal of Applied Systemic Studies (IJASS) 2024 Vol.11 No.2 pp.138 - 158 Abstract: COVID-19 impacted the overall economy and social order in any country from 2019, and Taiwan firstly setup a control centre which turned out to an excellent policy for the prevention and stemmed the spread of the disease by strengthening the publicity of patients' health to prevent the pandemic. Thus, the study is motivated to identify COVID-19 and Taiwan as research subjects. This study utilises the pandemic data (from January 2020 to May 2020) of five countries and proposes a hybrid time series-based method to analyse the diagnosis rates and mortality rates. Consequently, the USA, Russia, Spain, and Taiwan's forecast results fall within the confidence interval; Brazil's forecast results exceed the confidence interval. Despite the limitations, the proposed model can still be used as a viable alternative for predicting future pandemics. The empirical results of this study benefit researchers by avoiding the prodigality of medical resources from proper forecasting. Inderscience Publishers - linking academia, business and industry through research

Title: Identifying diagnosis and mortality of COVID-19 by learning a sequence-to-sequence ARIMA-based model

Authors: You-Shyang Chen; Jerome Chih Lung Chou; Naiying Hsu; Ting Yi Kuo

Addresses: College of Management, National Chin-Yi University of Technology, No. 57, Sec. 2, Zhongshan Rd., Taiping Dist., Taichung 411030, Taiwan ' School of Management, National Taiwan University of Science and Technology, No. 43, Keelung Rd., Sec. 4, Da'an Dist., Taipei City 106335, Taiwan ' Department of Information Management, Hwa Hsia University of Technology, 111, Gong Jhuan Rd., Chung Ho District, New Taipei City 235, Taiwan ' Department of Information Management, Hwa Hsia University of Technology, 111, Gong Jhuan Rd., Chung Ho District, New Taipei City 235, Taiwan

Abstract: COVID-19 impacted the overall economy and social order in any country from 2019, and Taiwan firstly setup a control centre which turned out to an excellent policy for the prevention and stemmed the spread of the disease by strengthening the publicity of patients' health to prevent the pandemic. Thus, the study is motivated to identify COVID-19 and Taiwan as research subjects. This study utilises the pandemic data (from January 2020 to May 2020) of five countries and proposes a hybrid time series-based method to analyse the diagnosis rates and mortality rates. Consequently, the USA, Russia, Spain, and Taiwan's forecast results fall within the confidence interval; Brazil's forecast results exceed the confidence interval. Despite the limitations, the proposed model can still be used as a viable alternative for predicting future pandemics. The empirical results of this study benefit researchers by avoiding the prodigality of medical resources from proper forecasting.

Keywords: diagnosis rate; mortality rate; new coronary pneumonia - COVID-19; time series forecasting; smoothing index; ARIMA model.

DOI: 10.1504/IJASS.2024.140024

International Journal of Applied Systemic Studies, 2024 Vol.11 No.2, pp.138 - 158

Accepted: 20 Feb 2024
Published online: 15 Jul 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: Identifying diagnosis and mortality of COVID-19 by learning a sequence-to-sequence ARIMA-based model

Keep up-to-date